Research Methods

HideShow resource information

Lab Experiments


  • An IV manipulated by the researcher in an attempt to produce a change in the DV.
  • High Control
  • Conducted in an artifical environment


  • Extraneous variables can be controlled - Know the cause of the change in th DV
  • Greater control allows ability to replicate study thus ensures reliability


  • Lacks ecological validity due to artifical environment
  • Higher risk of demand characteristics
  • May raise ethical issues of deception
  • Total control of all variables not possible
  • Results more likely to be biased by sampling
1 of 15

Field Experiments


  • IV is manipulated
  • Carried out in participant's natural environment


  • Carried out in natural environment - greater ecological validity
  • Less bias from sampling, demand characterics etc


  • More difficult to control all variables - may not be able to infer cause & effect
  • More difficult to replicate - may not be reliable
  • Ethical problems e.g. privacy, fully informed consent etc
2 of 15

Quasi Experiments


  • IV naturally exists e.g. age, gender etc
  • Mainly conducted in an artifical environment


  • As the condition already varies it allows comparison between different types of people


  • Hard to infer cause & effect as IV has not been directly manipulated
  • Random allocation to conditions is not possible - may cause confounding variables, may affect the internal validity
  • As they are often conducted in an artifical environment they have the same weaknesses as lab experiments
3 of 15

Natural Experiments


  • IV is changed through natural occurrence
  • Conducted in natural environment


  • High ecological validity
  • Little bias & demand characteristics


  • Hard to infer cause & effect due to little control
  • Very difficult to replicate
  • Ethical Issues e.g. privacy, consent etc
4 of 15

Correlational Analysis

Measures the relationship between two co-variables.

Postive Correlation: Whilst one variable increases so does the other

Negative Correlation: Whilst one variable increases the other decreases

Correlation coefficient: A number that shows the strength of the correlation, this is between -1 and +1

+1 = Perfect postive correlation

0 = No correlation

-1 = Perfect negative correlative

A scattergraph is used to show the correlation

5 of 15


Naturalistic Observation: Refers to the observation of behaviour in its natural setting - the reseracher makes no attempt to interfere with the behaviour

Controlled Observation: Some variables in the environment are controlled by the researcher

Overt: Aware of being observed

Covert: Unaware of being observed

Participant: Observer is part of the group being observed

Non-Partipiant: Observer watching/listening to the behaviour of others

6 of 15

Observations - Evaluation 1

In General

  • Tests existing findings
  • Meanings - can describe behaviour but are unlikely to tell us why
  • Observer Bias
  • Can be used in situations where it would be unethical to manipulate variables

Naturalistic observations

  • Realism
  • Lack of control

Controlled observations

  • Lack of realism
  • High control
7 of 15

Observations - Evaluation 2

Covert observations

  • Low demand characteristics and observer effects
  • Ethical Issues - lack of informed consent

Overt observations

  • Risk of demand characteristics and observer effects
  • Lack of ethical issues

Participant observations

  • Greater insights into behaviour
  • Objectivity of the observations are effected by being part of the group/situation

Non-Participant observations

  • Lack of direct involvement
  • Data lacks the richness
8 of 15


Closed questions: Questions where the participant gets a range of answers selected by the researcher

Open questions: Questions where the participants are not restricted by a range of answers and can expand on their answers


  • Cheap & Quick
  • Wide range of data - Quantitative and Qualitative
  • Ease of analysis


  • Social Desirability Bias
  • Researcher Effects
  • Difficult to control
  • Different interpretations of questions
9 of 15

Questionnaire Design

Questions to avoid:

  • Leading questions
  • Ambiguous questions
  • Emotive questions
  • Double-barelled questions
  • Jargon
  • Negatives
  • Complex questions
  • Impossible questions

Questions should try to include

  • Filler questions
  • Easy questions
  • Lie detection questions
  • A balance of yes/no responses - Tendency to answer Yes
10 of 15

Research Designs - Repeated Group Design

Repeated Group Design: Participants are used in both conditions


  • Participant Variables are kept constant
  • More powerful statistics are used
  • Fewer participants are needed


  • Order effects - can be avoided by counterbalancing
  • Demand Characteristics
11 of 15

Research Designs - Individual Group Design

Individual Group Design: Different participants used in each condition


  • Order effects do not occur
  • Demand characteristics less of a problem
  • The same test can be used in each condition


  • Participant variables are different - confounding variables
  • Less powerful statistical tests
  • More participants are required
12 of 15

Research Designs - Matched Pairs Design

Matched Pairs Design: Different but similar participants are used in each condition


  • Participant variables kept constant
  • No order effects
  • Less Demand Characteristics
  • The same test can be used
  • More powerful statistical tests


  • Participant variables can never be matched in all aspects
  • Matching participants is time consuming and difficult
  • More participants are required
13 of 15

Sampling 1

Random: Every member of target pop. has an equal chance of being selected

  • Provides the best chance of an unbiased representative sample
  • Compiling a list of everyone in the target pop. is impractical - time consuming, expensive

Stratified: Dividing the target pop. into important subcategories and then selects members of the categories in the correct proportions

  • A deliberate effort is made to identify the characteristics of a sample for it to be representative
  • It can be difficult to identify important characteristics & very time consuming

Opportunity: Selecting participants that are available at the time

  • It is quick, convenient and economical compared to the other methods
  • Unrepresentative and often biased on the part of the researcher to pick helpful participants
14 of 15

Sampling 2

Volunteer:Individuals who have determined their involvement in a study

  • Convenient and if volunteering is made in terms of informed consent then it is ethical
  • Unrepresentative samples - each volunteer is different

Systematic: A sample obtained by selecting every nth person on a list. The numerical interval is to be applied consistently.

  • Unbiased as Ps are selected by an objective system
  • Not truly unbiased/random unless number is selected through random method
15 of 15


No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Research methods and techniques resources »